ABSTRACT
PURPOSE: To develop consensus regarding clinical disease severity classification systems for diabetic retinopathy and diabetic macular edema that can be used around the world, and to improve communication and coordination of care among physicians who care for patients with diabetes. DESIGN: Report regarding the development of clinical diabetic retinopathy disease severity scales. PARTICIPANTS: A group of 31 individuals from 16 countries, representing comprehensive ophthalmology, retina subspecialties, endocrinology, and epidemiology. METHODS: An initial clinical classification system, based on the Early Treatment Diabetic Retinopathy Study and the Wisconsin Epidemiologic Study of Diabetic Retinopathy publications, was circulated to the group in advance of a workshop. Each member reviewed this using e-mail, and a modified Delphi system was used to stratify responses. At a later workshop, separate systems for diabetic retinopathy and macular edema were developed. These were then reevaluated by group members, and the modified Delphi system was again used to measure degrees of agreement. MAIN OUTCOME MEASURES: Consensus regarding specific classification systems was achieved. RESULTS: A five-stage disease severity classification for diabetic retinopathy includes three stages of low risk, a fourth stage of severe nonproliferative retinopathy, and a fifth stage of proliferative retinopathy. Diabetic macular edema is classified as apparently present or apparently absent. If training and equipment allow the screener to make a valid decision, macular edema is further categorized as a function of its distance from the central macula. CONCLUSIONS: There seems to be a genuine need for consistent international clinical classification systems for diabetic retinopathy and diabetic macular edema that are supported with solid evidence. The proposed clinical classification systems provide a means of appropriately categorizing diabetic retinopathy and macular edema. It is hoped that these systems will be valuable in improving both screening of individuals with diabetes and communication and discussion among individuals caring for these patients.
